We will investigate the cognitive organization of word processing, verbal short-term memory (STM) and word learning. These three capacities are fundamental to learning, and their integrity has implications for rehabilitation of neurologically-based language deficits. We propose to continue our studies of the effects of semantic and phonological impairment on language, STM and learning abilities of individuals with aphasia and will extend these analyses to individuals with semantic dementia. Our focus will be on the detrimental effects of semantic impairments on learning and response to treatment in these populations. Semantic impairments are present in both semantic dementia and aphasia but are manifested differently in the two populations. These differences are as yet not fully understood but are broadly characterized as semantic access vs. semantic knowledge impairments. Our goal is to elucidate the behavioral and neuroanatomical features of semantic access and knowledge disorders and the effects of each on word processing, STM and learning. We will obtain behavioral and neuroimaging data (structural MRI scans) on four groups defined by the etiology of their impairment (aphasia, semantic dementia) and locus of disruption within a psycholinguistic model of word processing. Additionally, we will identify factors that stress semantic short- term working memory to test the hypothesis that this factor is at the root of semantic access deficits and that semantic STM deficits are associated with impaired executive functions. In a second project, we will investigate effects of semantic and phonological impairment in aphasia on learning to recognize and produce new words in order to determine how these impairments affect this process. This project is linked to a third ongoing project that investigates priming treatments for word retrieval disorders in aphasia. Here we aim to test the efficacy of a new version of our contextual priming treatment that targets output pathways of naming directly. The cognitive data gathered in these three projects will be brought to bear on the ongioing development of a computational model of verbal STM, word processing and learning.